Chapter 1:
Introduction and overview
2
Diverse combinatorial protein libraries have the potential to generate molecules which
can bind virtually any desirable target. The number of potential protein targets is
enormous; including post translational modifications and alternative splicing, the human
proteome contains millions of proteins (1). Including microbial proteins, the number of
useful targets is even greater. The natural diversity represented by the mammalian
immune system has long been exploited for the generation of novel affinity reagents (2).
In vitro selection techniques such as phage display, ribosome display, and mRNA display
have been developed for generation of novel protein affinity reagents as an alternative to
animal immunization (3). With the advent of these techniques, synthetic and semi- synthetic immunoglobulin libraries have been developed (4-6). One drawback of using
antibodies as affinity reagents, however, is that antibody immunoglobulin domains
require intrachain disulfide bonds for stability (7). This requirement makes antibodies
difficult to express in large quantities and also limits their utility for intracellular
applications. In conjunction with the development of in vitro selection techniques, there
has been interest in the generation of alternative protein scaffolds which, like antibodies,
are able to tolerate sizable sequence diversity but, unlike antibodies, also express well in
the intracellular environment (3, 8, 9).
There are many applications for the molecular recognition of natural proteins by
novel protein affinity reagents (10). Recombinant antibodies are becoming increasingly
common as therapeutics (11). Other applications include standard laboratory techniques
such as immunoprecipitation, ELISA, western blot, and immunofluorescence
microscopy. Applications for protein detection are expanding with the development of 3
ultra-sensitive nano-electonic devices (12) and the use of engineered antibodies in nuclear medicine (13).
Despite difficulties in the expression of antibody fragments, a number of applications for intracellular recognition of protein targets have been demonstrated by intracellular antibodies (intrabodies) (14). For example, protein affinity reagents can be used as tools for intracellular detection and visualization (15). Exogenous binders
expressed as fluorescent protein fusions enable the visualization of dynamic proteins in
live cells. This application also allows the detection of protein specific to particular
conformational states or post translational modifications.
Novel protein affinity reagents may also be useful for functional genomics by the
direct inhibition of protein activity inside the cell. This technique was first demonstrated
in yeast by the expression of an antibody which limited the activity of alcohol
dehydrogenase (16). Inhibitory proteins can be selected to target specific protein domains, conformations, or modifications, whereas both gene knock-out and RNAi
techniques are not specific. Many intrabodies have been reported to inhibit proteins
important in cancer development which are considered “undrugable” (17). However, the
application for intrabodies as therapeutics is dependent on the progress of gene therapy
technology.
One industrial application that does not face gene delivery obstacles is the
generation of transgenic plants that express intrabodies. Intrabodies have been generated
that confer resistance against disease (18, 19) and improve the metabolic properties of plants (20). In addition to these reports, there are also many examples of plant intrabodies that were developed for functional genomics experiments as an alternative to gene 4
silencing techniques (21). For example, an antibody that inhibits heat shock protein oligomerization in vivo was used to illustrate the functional importance of this protein where previous genetic techniques proved inconclusive (22).
These reports demonstrate the potential for using intrabodies to determine protein function. However, because many antibody fragments are not stable in vivo, the usable
diversity of antibody libraries may only represent a fraction of the total diversity. In some
cases, very few or even no antibody fragments are able to interact with the intended target
in vivo (18, 23). In order to improve the probability of obtaining molecules functional in
vivo, screens based on the two-hybrid method have been implemented after partial
enrichment of complex antibody libraries (23, 24). Also, many intrabodies that are
functional in vivo have the propensity to form aggregates (25, 26). The ability to bind and
precipitate a protein target is desirable for certain therapeutic applications, but is not
desirable for in vivo detection and functional genomics. While aggregation ensures
inhibition by effectively removing the protein target from the cytosol, the aggregates may
adversely affect the cell. For example, these types of aggregates have been shown to
disrupt the ubiquitin dependent proteasomal degradation pathway and result in a
propensity to induce apoptosis (27, 28).
There have been successful attempts to engineer antibody fragments that are more
stable in vivo (29-31). Alternatively, non-immunoglobulin domains have been used as
scaffolds for combinatorial libraries which may improve upon antibody libraries (32).
One successful protein library is based on the ankyrin domain (8, 33). This library has been used in ribosome display selection experiments to generate novel affinity reagents that are functional in vivo (34, 35). One ribosome display selection yielded molecules 5 that were able to inhibit a tobacco etch virus proteinase in vivo, and stable expression of the ankyrin inhibitors in plants could lead to viral resistance (36).
The ankyrin library was created by consensus design where non-conserved residues were randomized throughout the protein surface (8). We took an alternative approach and utilized the fibronectin type III domain to create a combinatorial protein library with diversity localized within two adjacent randomized loops (9). The library scaffold was based on the 10th fibronectin type III domain of human fibronectin, originally implemented by Koide et al. for phage display selections (37). This domain is topologically analogous to the immunoglobulin fold; however, unlike immunoglobulin variable domains, it does not contain disulfide bonds and is able to be expressed at high levels in bacteria. Chapter 2 describes the design and construction of our library as well as expression and stability analysis of representative library members. Over half of the molecules encoded in the library are able to be expressed in bacteria. The domain was also demonstrated to tolerate mutations at the 17 randomized loop positions as four representative variants were shown to be structurally stable. In addition to demonstrating the utility of this scaffold for selection techniques, this report illustrates the vast quantity of sequence space that is accessible for both natural and directed evolution of novel functions.
Our fibronectin-based combinatorial library was designed for in vitro selection by mRNA display (38). mRNA display technology is an elegant and simple method for the generation of novel high-affinity peptide and protein binders to specific protein targets
(39). The utility of the technique lies in the ability to link a unique protein phenotype to its genotype via a covalent bond. This is achieved by splint-mediated ligation of an 6
oligonucleotide bearing a 3’ puromycin to an mRNA pool synthesized by in vitro
transcription of the DNA that encodes the combinatorial protein library (Figure 1.1,
panel A). The mRNA library is then translated in vitro and fusion of the nascent peptide
chain to the mRNA is catalyzed by the ribosome when the ribosome stalls at the poly-dA
linker (Figure 1.1, panel B). This selection technique, which is entirely in vitro, has
advantages over other selection techniques including phage display and yeast display in that higher complexity libraries, over 10 trillion unique molecules, are accessible. Also the mRNA display format is monovalent. Since a covalent bond links the fusions, this selection technique also has an advantage in that selections can be performed at any desired level of stringency.
One of the benefits for using in vitro selected protein affinity reagents for proteomics or functional genomics is the ability to generate molecules that bind in a domain, conformation, or post translational modification-specific manner. To highlight the utility of our fibronectin library, we sought to generate binders that modulate and detect one of the most important pathways in the cell, the NF-κB pathway. NF-κB
proteins are ubiquitous transcription factors primarily involved in activation of pro-
inflammatory genes (40). The NF-κB pathway also plays an important role in cell survival, as well as in neuronal signaling (41). Activation of the classical NF-κB pathway
is controlled by ubiquitin dependent proteasomal degradation of three inhibitory proteins,
IκBα, β, and ε (42). These proteins are phosphorylated by IKK at two serines within a
conserved DSGXXS destruction motif that is recognized by the SCF-βTrCP E3 ligase
when phosphorylated (43). Chapter 3 describes a selection for fibronectin molecules that
specifically recognize the phosphorylated state of IκBα. We were able to evolve a 7 fibronectin molecule, labeled 10C17C25, that is able to discriminate between the phosphorylated and unphosphorylated states of IκBα with over 1000-fold specificity, measured by surface plasmon resonance. 10C17C25 was also able to recognize IκBα inside human kidney cells (293T) and inhibit its degradation. 10C17C25 was able to pull- down IκBα only in cells in which the NF-κB pathway was activated. We have also demonstrated for the first time the application of novel protein affinity reagents for use in
FRET sensors of kinase activity. Our IKK FRET sensors are similar to previous sensors which rely on natural phospho-specific binding domains that detect kinase activity in cells reversibly in real-time without disruption of the pathway being detected (44, 45).
With the growing threat of epidemics similar to the recent SARS outbreak, novel tools to detect and study viruses are needed. Many intrabodies have been described which modulate viral proteins (17). These experiments validate the potential of viral proteins as targets for antiviral therapies. Direct inhibition of viral proteins in cells infected with unmodified virus is a useful tool for neutralization of viral genes for functional analysis where gene knock-outs or RNAi techniques are not applicable (25, 46). We sought to demonstrate the potential for using SARS nucleocapsid-binding fibronectins to detect and probe viral protein function in a domain-specific manner. Chapter 4 describes a selection which generated molecules that bind nucleocaspsid (N) protein with low nanomolar affinity after only 6 rounds of enrichment. The primary function of N is to package the
RNA genome within the viral envelope. However, N is found in abundant levels in the serum of infected patients and represents a potential target for early detection of SARS infection. In addition to validating potential therapeutic strategies, these molecules may be used to better establish the many additional roles coronavirus N protein plays in viral 8 replication and in host-cell interactions (47). We were able to demonstrate over 1000-fold inhibition of viral production by intracellular expression of the most potent SARS- inhibiting fibronectin. Also, as we were able to obtain binders to non-overlapping epitopes, we were able to demonstrate the value of synergistic inhibition by targeting two domains within the same protein. Although most antiviral therapeutic targets are enzymes, we demonstrated the efficacy of inhibiting the protein interactions mediated by a virus structural protein.
Finally, Appendix A describes the vectors used for the expression of fibronectins.
Vectors were created for evolving protein stability, expressing protein in both bacteria and mammalian cell culture, and for cloning selection products into a general FRET sensor vector. Appendix A illustrates the ORF of each vector used, describes the rationale for creating the vector, and describes the method for vector construction.
9
References (1) Jensen, O. N. (2004) Modification-specific proteomics: characterization of post- translational modifications by mass spectrometry. Curr Opin Chem Biol 8, 33-41. (2) Margulies, D. H. (2005) Monoclonal antibodies: producing magic bullets by somatic cell hybridization. J Immunol 174, 2451-2. (3) Lipovsek, D., and Pluckthun, A. (2004) In-vitro protein evolution by ribosome display and mRNA display. J Immunol Methods 290, 51-67. (4) Desiderio, A., Franconi, R., Lopez, M., Villani, M. E., Viti, F., Chiaraluce, R., Consalvi, V., Neri, D., and Benvenuto, E. (2001) A semi-synthetic repertoire of intrinsically stable antibody fragments derived from a single-framework scaffold. J Mol Biol 310, 603-15. (5) Knappik, A., Ge, L., Honegger, A., Pack, P., Fischer, M., Wellnhofer, G., Hoess, A., Wolle, J., Pluckthun, A., and Virnekas, B. (2000) Fully synthetic human combinatorial antibody libraries (HuCAL) based on modular consensus frameworks and CDRs randomized with trinucleotides. J Mol Biol 296, 57-86. (6) Maynard, J., and Georgiou, G. (2000) Antibody engineering. Annu Rev Biomed Eng 2, 339-76. (7) Ewert, S., Honegger, A., and Pluckthun, A. (2004) Stability improvement of antibodies for extracellular and intracellular applications: CDR grafting to stable frameworks and structure-based framework engineering. Methods 34, 184-99. (8) Binz, H. K., Stumpp, M. T., Forrer, P., Amstutz, P., and Pluckthun, A. (2003) Designing repeat proteins: well-expressed, soluble and stable proteins from combinatorial libraries of consensus ankyrin repeat proteins. J. Mol. Biol. 332, 489-503. (9) Olson, C. A., and Roberts, R. W. (2007) Design, expression, and stability of a diverse protein library based on the human fibronectin type III domain. Protein Sci 16, 476-84. (10) Stoevesandt, O., and Taussig, M. J. (2007) Affinity reagent resources for human proteome detection: initiatives and perspectives. Proteomics 7, 2738-50. (11) Carter, P. J. (2006) Potent antibody therapeutics by design. Nat Rev Immunol 6, 343-57. 10
(12) Patolsky, F., Zheng, G., and Lieber, C. M. (2006) Nanowire sensors for medicine and the life sciences. Nanomed 1, 51-65. (13) Kenanova, V., and Wu, A. M. (2006) Tailoring antibodies for radionuclide delivery. Expert Opin Drug Deliv 3, 53-70. (14) Stocks, M. (2005) Intrabodies as drug discovery tools and therapeutics. Curr Opin Chem Biol 9, 359-65. (15) Nizak, C., Monier, S., del Nery, E., Moutel, S., Goud, B., and Perez, F. (2003) Recombinant antibodies to the small GTPase Rab6 as conformation sensors. Science 300, 984-7. (16) Carlson, J. R. (1988) A new means of inducibly inactivating a cellular protein. Mol Cell Biol 8, 2638-46. (17) Kontermann, R. E. (2004) Intrabodies as therapeutic agents. Methods 34, 163-70. (18) Boonrod, K., Galetzka, D., Nagy, P. D., Conrad, U., and Krczal, G. (2004) Single- chain antibodies against a plant viral RNA-dependent RNA polymerase confer virus resistance. Nat Biotechnol 22, 856-62. (19) Tavladoraki, P., Benvenuto, E., Trinca, S., De Martinis, D., Cattaneo, A., and Galeffi, P. (1993) Transgenic plants expressing a functional single-chain Fv antibody are specifically protected from virus attack. Nature 366, 469-72. (20) Jobling, S. A., Jarman, C., Teh, M. M., Holmberg, N., Blake, C., and Verhoeyen, M. E. (2003) Immunomodulation of enzyme function in plants by single-domain antibody fragments. Nat Biotechnol 21, 77-80. (21) Nolke, G., Fischer, R., and Schillberg, S. (2006) Antibody-based metabolic engineering in plants. J Biotechnol 124, 271-83. (22) Miroshnichenko, S., Tripp, J., Nieden, U., Neumann, D., Conrad, U., and Manteuffel, R. (2005) Immunomodulation of function of small heat shock proteins prevents their assembly into heat stress granules and results in cell death at sublethal temperatures. Plant J 41, 269-81. (23) Visintin, M., Tse, E., Axelson, H., Rabbitts, T. H., and Cattaneo, A. (1999) Selection of antibodies for intracellular function using a two-hybrid in vivo system. Proc. Natl. Acad. Sci. USA 96, 11723-8. 11
(24) Tse, E., Lobato, M. N., Forster, A., Tanaka, T., Chung, G. T., and Rabbitts, T. H. (2002) Intracellular antibody capture technology: application to selection of intracellular antibodies recognising the BCR-ABL oncogenic protein. J. Mol. Biol. 317, 85-94. (25) Vascotto, F., Campagna, M., Visintin, M., Cattaneo, A., and Burrone, O. R. (2004) Effects of intrabodies specific for rotavirus NSP5 during the virus replicative cycle. J Gen Virol 85, 3285-90. (26) Cardinale, A., Filesi, I., and Biocca, S. (2001) Aggresome formation by anti-Ras intracellular scFv fragments. The fate of the antigen-antibody complex. Eur J Biochem 268, 268-77. (27) Kopito, R. R., and Sitia, R. (2000) Aggresomes and Russell bodies. Symptoms of cellular indigestion? EMBO Rep 1, 225-31. (28) Cardinale, A., Filesi, I., Mattei, S., and Biocca, S. (2003) Evidence for proteasome dysfunction in cytotoxicity mediated by anti-Ras intracellular antibodies. Eur J Biochem 270, 3389-97. (29) der Maur, A. A., Zahnd, C., Fischer, F., Spinelli, S., Honegger, A., Cambillau, C., Escher, D., Pluckthun, A., and Barberis, A. (2002) Direct in vivo screening of intrabody libraries constructed on a highly stable single-chain framework. J. Biol. Chem. 277, 45075-85. (30) Tanaka, T., Lobato, M. N., and Rabbitts, T. H. (2003) Single domain intracellular antibodies: a minimal fragment for direct in vivo selection of antigen-specific intrabodies. J Mol Biol 331, 1109-20. (31) Tanaka, T., and Rabbitts, T. H. (2003) Intrabodies based on intracellular capture frameworks that bind the RAS protein with high affinity and impair oncogenic transformation. EMBO J. 22, 1025-35. (32) Binz, H. K., Amstutz, P., and Pluckthun, A. (2005) Engineering novel binding proteins from nonimmunoglobulin domains. Nat Biotechnol 23, 1257-68. (33) Binz, H. K., Amstutz, P., Kohl, A., Stumpp, M. T., Briand, C., Forrer, P., Grutter, M. G., and Pluckthun, A. (2004) High-affinity binders selected from designed ankyrin repeat protein libraries. Nat. Biotechnol. 22, 575-82. 12
(34) Amstutz, P., Binz, H. K., Parizek, P., Stumpp, M. T., Kohl, A., Grutter, M. G., Forrer, P., and Pluckthun, A. (2005) Intracellular kinase inhibitors selected from combinatorial libraries of designed ankyrin repeat proteins. J Biol Chem 280, 24715-22. (35) Amstutz, P., Koch, H., Binz, H. K., Deuber, S. A., and Pluckthun, A. (2006) Rapid selection of specific MAP kinase-binders from designed ankyrin repeat protein libraries. Protein Eng Des Sel 19, 219-29. (36) Kawe, M., Forrer, P., Amstutz, P., and Pluckthun, A. (2006) Isolation of intracellular proteinase inhibitors derived from designed ankyrin repeat proteins by genetic screening. J Biol Chem 281, 40252-63. (37) Koide, A., Bailey, C. W., Huang, X., and Koide, S. (1998) The fibronectin type III domain as a scaffold for novel binding proteins. J Mol Biol 284, 1141-51. (38) Roberts, R. W., and Szostak, J. W. (1997) RNA-peptide fusions for the in vitro selection of peptides and proteins. Proc Natl Acad Sci U S A 94, 12297-302. (39) Takahashi, T. T., Austin, R. J., and Roberts, R. W. (2003) mRNA display: ligand discovery, interaction analysis and beyond. Trends Biochem Sci 28, 159-65. (40) Viatour, P., Merville, M. P., Bours, V., and Chariot, A. (2005) Phosphorylation of NF-kappaB and IkappaB proteins: implications in cancer and inflammation. Trends Biochem Sci 30, 43-52. (41) Meffert, M. K., and Baltimore, D. (2005) Physiological functions for brain NF- kappaB. Trends Neurosci 28, 37-43. (42) Pomerantz, J. L., and Baltimore, D. (2002) Two pathways to NF-kappaB. Mol Cell 10, 693-5. (43) Karin, M., and Ben-Neriah, Y. (2000) Phosphorylation meets ubiquitination: the control of NF-[kappa]B activity. Annu Rev Immunol 18, 621-63. (44) Sato, M., Ozawa, T., Inukai, K., Asano, T., and Umezawa, Y. (2002) Fluorescent indicators for imaging protein phosphorylation in single living cells. Nat Biotechnol 20, 287-94. (45) Wang, Y., Botvinick, E. L., Zhao, Y., Berns, M. W., Usami, S., Tsien, R. Y., and Chien, S. (2005) Visualizing the mechanical activation of Src. Nature 434, 1040- 5. 13
(46) Corte-Real, S., Collins, C., Aires da Silva, F., Simas, J. P., Barbas, C. F., 3rd, Chang, Y., Moore, P., and Goncalves, J. (2005) Intrabodies targeting the Kaposi sarcoma-associated herpesvirus latency antigen inhibit viral persistence in lymphoma cells. Blood 106, 3797-802. (47) Enjuanes, L., Almazan, F., Sola, I., and Zuniga, S. (2006) Biochemical aspects of coronavirus replication and virus-host interaction. Annu Rev Microbiol 60, 211- 30.
14
Oligonucleotide A PCR DNA Transcription mRNA
Purification P Ligation P
B
Figure 1.1 mRNA display library synthesis and fusion formation. A) Cartoon depicting synthesis of an mRNA pool with a 3’ puromycin for mRNA display. Simple libraries are encoded by single randomized sequence (red). Constant regions are depicted in black. The 5’ constant region includes the T7 promoter sequence, a translation enhancer sequence, and the initiator methionine codon. The 3’ constant region provides sequence for primer annealing and splint hybridization and also encodes a flexible linker for spacing the nucleic acid fusion. B) Cartoon depicting in vitro translation and fusion formation.